Simon Wyndham

With a low budget shoot coming up in a forest soon where there can be extremes of dappled light going on in the background, I was wondering if anyone here uses the Ultra con, or digicon filters at all, and whether or not the concensus was that they were valuable filters or not?

Peter, there is an example of a silent movie look I did for a project on an XM1 near the end of my reel http://www.simonwyndham.co.uk/reel.htm I used FilmFX 2.35 using a custom setting I based upon Buster Keatons "The General". I got the speed up effect using After Effects time remapping, though I forget how many fps I based things on.

From what I know the Dionic 90's they will not actually allow themselves to get to a critical discharge point. I believe that they have an automatic protection circuit to prevent this happening. Anyone know anything about this?

I'm not so sure. The minds of electronics giants are hard to figure out! Certainly replication of certain gamma characteristics would not spell the end of film. The cameras are standard def after all, and the actual full dynamic range of film can not be captured despite the wide capability of the camera head. It would seem odd to me that the numbers mentioned would correspond to specific stocks if that was not what they were attempting to replicate. Remember we are talking about gamma characteristics here, not actual full replication. It would not surprise me at all if most of these settings were just placed into the camera purely because they can, possibly as an afterthought. They may well have added in a generic film gamma setting along the lines of the SDX and Varicam, and then just added the others for the hell of it. I don't think there is any need to read too much into it. But as I mentioned I am awaiting a response from Sony as to what the settings are meant for and whether they do represent the specific stocks gamma characteristics. If they do, it may seem pointless, but that would be a mute point as I am interested in knowing how similar these gamma selections are to the Panny Film gamma settings as well finding out exactly what the gamma characteristics of the settings are. Just my curiosity to find out exactly how things work and why is all.

It could just be a case of they could, so they did. You are correct, there is minimal differences between most of those settings. Although from a general use point of view, they do handle the highlights far better than the STD gamma tables. So in that respect I think they *should* be useful even if there is only a need for one of them. I am trying to find out some clarification from Sony however.

Hmm, interesting. From an official Sony XDCAM rep I was told that the film gamma tables represented the following. Maybe I got the wrong end of the stick with regards to the numbers? 1. Average (Film2-5) 2. Film No. 5245 3. Film No. 5248 4. Film No. 5293 5. Film No. 5296 When selected on the camera, number 5 gives the more contrasty picture (as in difference between light and dark). The others appear to give much more contrast in terms of latitude. I'll try to find out some more about them. Specifically I woiuld be interested in knowing if they were designed purely for the purposes of filmout.

Interesting. Have you had experience of the 950? The PDW series of cameras, and I assume the newer 400 have 5 film gamma settings supposedly repllcating the following Kodak stocks; 1. Standard Film (average of all below) 2. Kodak 50D 3. Kodak Eastman EXR 100T 4. Kodak Eastman EXR 200T 5. Kodak Eastman EXR 500T Anyone who uses the SDX any comments on its gamma presets and how often people use them?

Hi David, Yes, there are quite a number of people who suggest that for reasons that I too am completely baffled by. It is amazing how many people do not understand what the progressive mode on their cameras does. Of course these people are mostly general users rather than people who do this for a living. I may ask Alan about those points regardless as a general enquiry as I too would be interested in how he accounts for those. But as you mentioned previously, much of the time we should try and live with the limitations and work within them rather than chasing something that may not be there. Incidentally, does the F900 and the SDX have film style gamma settings? And do you ever use them?

Hi David, Alan hangs out at the UK based DV Doctor forums in the A/V Hardware section. http://forums.dvdoctor.net/ It would be well worth popping over there to ask him any questions you may have about his research and setups as I am sure that the information given will be of benefit to a good many people. He's a very approachable guy. Besides, it would be nice to have some more 'names' over there :)

Yes. This is the main point about Alan Roberts settings. Obtaining the filmlike motion is more than progressive scan because it also depends on the differences in the ways film handles the different frequencies compared to video. One of the largets parts of Alans adjustments involve making the motion characteristic of progressive HD video more match that of film by adjusting the frequency handling. Thats how I understand it anyhow. :) So, as you rightly point out, the motion characteristic is the biggest part of the 'filmlook'. But out of the box progressive video isn't quite the same in the way it appears to move. In this respect Alans settings are universally useful no matter what look one is going for. But this kind of adjustment really needs someone who knows what they are doing. The BBC White Papers by Alan are very informative on this very subject. http://www.bbc.co.uk/rd/pubs/whp/whp053.shtml http://www.bbc.co.uk/rd/pubs/whp/whp034.shtml Warning, the second paper will make your head explode!

Certainly one should make an effort to understand exactly what the settings are doing. But it is not an easy thing for someone without the right background to try and replicate what somebody who has made it their function in life to research these sorts of settings. People may want to create their own looks for the camera. But it helps to know that the camera has the best baseline from which to work from as possible. At least then, depending on what you have asked the engineer to do, you will have one setting that you know makes you happy and works in the majority of everyday situations. Yes, lighting is a big factor. But I might add that even here there are cinematographers who like to shoot in natural light. and with landscape shots, and places where there is no avoiding the high contrast in the background, I know I would personally be happier knowing that my camera is going to cope as best as the hardware allows, not just by guessing a setting, but by *knowing* a setting/s. I am with you totally on avoiding digital enhancements as much as possible. This is a very good reason why I use setting, again that were researched using the right equipment, that reduce such things. Only just now I have been playing around with turning the knee functions off completely to see the effect. I may try working with it turned off for a while to let the highlights blow out more naturally as I found the knee circuits, more often than not, were muddying the highlights. At the same time I found out just how much more latitude in the highlights switching the camera to FLM gamma gave. Compared to the standard video style setting the skies would blow out completely, whereas the FLM gamma kept the nice blues and cloud definition while also giving me a tad more shadow detail. But rather than being about how I think the picture looks aesthetically, this is more a case of trying to get the camera to give me as much as it can.

David, thats the beauty of Alans research. His settings aim to give the maximum dynamic range, filmlike highlight roll off, minimal artefacts/noise etc as possible. He's done a lot of research into the way film looks during movement. A big part of Alans settings, as MDO pointed out, is to replicate the frequency response of film, and therefore reduce the 'judder' that usually comes from straight progressive video with no tweaking due to the way the human eye perceives motion. Alans can set up a camera to replictae whatever you need it to. But his base settings aim to replicate neg film, as well as giving as much range as possible so the picture can be worked with in post. Processes such as skip bleach are a process after shooting has taken place as you know. As such Alans settings are ideal for this kind of effect replication in post as it should be. Since the maximum dynamic range is available that the camera is capable of, the skip-bleach effect can be tweaked to perfection rather than being a hit and miss affair should the effect be programmed into the cameras menus. Same goes for cross processing and colour reversal. While with film cameras you have to live with what those types of film stocks give you at the end of the day, with digital there is no point in performing such actions in camera. Why risk it? Better to let the camera give you as much picture info as possible and then tweak it later. The default settings on most cameras are also far too sharp. Blow up default settings onto a larger screen and edge enhancement sticks out like a sore thumb. I prefer the picture to have minimal digital edge enhancement. Now as to why the manufacturers do not place these settings in camera as default settings, some of them do to a degree. The film gamma settings for filmout on a Varicam for example, the FLM gamma settings on the XDCAMs etc. They go a long way towards this. Now Alans settings will look very flat. A camera manufacturer generally wants their camera to look colourful, very sharp and bold. So many of the default settings on cameras are set waaay too high. If you want edge enhancement it is better to do this in post too where you will have a much more precise control over the line widths etc. Now while it is possible for people to try messing around with the setting themselves, what guys like Alan Roberts do is much more precise and in depth, and is the result of years of specific research into exactly what they are doing. It can take 2 days or so to acheive the final settings on an unfamiliar camera with all the relevant test equipment, signals, and colour charts to work from. What Alan does is precisely measured. Its a testement to his work that the BBC uses his settings as default, and believe me, the Beeb have the absolute highest standards when it comes to camera setup.

I had thought about that in the past, but the menus and camera heads between cameras are vastly different. The settings for the 790 and 700 would not work effectively on, say, a 510 or the SDX. People do tend to be quite protective about settings for cameras. I think Alan is the best bet for having the camera set up correctly. He seems to be the UK equivilent of Roger Macie. The filmlook such as the BBC settings being discussed are something that really needs a thoroughly experience engineer with various monitors and colour charts to sort out.

Yep, the settings researched and devised by Alan Roberts I believe. I believe he still does some freelance work too to help people set up their cameras for filmlook. His settings apparently aim to set up the camera for a much more filmlike response in the various frequencies as well as increasing dynamic range to the max. I've been desperate to find out what settings the BBC use for XDCAM, but alas one has to actually be working for the Beeb on a project to obtain them. :( The only settings I use for my 510 are the ones recommended by Swiss Effects, a film transfer house. http://www.swisseffects.ch They keep settings handy for many different cameras, and despite not using filmout I tend to use their detail settings all the time as they drastically reduce edge enhancement detail while not making the picture overly soft. I'm sure if anyone who is interested in this contacted them they would give you their settings for the SDX too.

Yes, thats a given.

In light of several articles regaring HD shooting stating that DCC should always be turned on, I have also found some opinion that states that they always use DCC (Dynamic Contrast Control) with caution due to its automation and 'hunting' effects that can sometimes come with it. What is the general concensus here regarding DCC? Would you only recommend it for highly controlled lighting conditions, or would you risk it in more unpredictable situations?

The camera is in interlaced mode by default. You need to adjust the menus. Go to the OPERATION menu, and scroll down to page 20 "FORMAT". Go to SCAN MODE and change it to PsF. Exit the menu. Sony recommend that you then turn the camera off for 10 seconds and then turn it on again after doing this. You will now be in PsF mode. But you need to switch the shutter on and set it to 1/50 otherwise things will look a little dreamy in motion. Hope this helps. :)

I can't think of any other things at the moment, but I'll let you know if I come up with any :)

No probs. I would like to expand on one thing Lars mentioned. Obviously whether or not you shoot in progressive (PsF) mode or interlaced mode depends on your director and producer. However if the descision is made to shoot in progressive scan don't forget to turn the shutter ON and switch it to 1/50 (or 1/48 if you are in NTSC land) otherwise you will get a very dreamy motioned picture! Having said that however, if you are in lower light levels and there isn't much motion in in the shots, by turning the shutter off (1/24 or 1/24) you will gain an extra stop of light. This can be very handy indeed, and is in fact what Michael Mann did when he shot Collateral and the night scenes in Ali when he shot on high def. With regards to the filters, I usually have my white balance offset slightly. Instead of using 5600k I set the preset A to 6300 for daylight as Sony's are well known to be slightly on the cool side. It's also handy to set the rear XLR sockets to automatic detection.

I own a 510. Great camera. Lars, interesting what you say about the flare setting. I'll have to try that out. Generally with the detail settings you can turn the detail way down. No focussing problems. Remember that '0' on the detail setting is still enhanced. Even if you turn the detail settings off completely you still don't actually lose detail. By turning up the peaking on the viewfinder you are adding the edge enhancement that would otherwise be added to the recorded picture to help with focussing. I prefer to have all my recorded pictures without edge enhancement lines. Usually I use the detail settings recommended by Swiss Effects as they minimise edge enhancement lines. They are; Aperture Level -15 Dtl h/v Ratio +10 Crispening -10 Master Black Gamma + 20 White Clip 109.5 Detail Freq +40 Detail - 40 The black gamma setting and white clip can be ignored if you are not doing a film transfer. I think that the built in FLM gamma settings would do a better job at increasing dynamic range than just using black stretch anyway. As for tips actually using the camera, go into the menus and set up the user buttons on the camera. I usually have the two on the top of the handle set to RET and Delete so that I can quickly delete the last clip if it was no good. Another tip is that when you shoot the first shot of a scene, go to the thumbnails menu and add that clip to a clip list. Save the cliplist with a relevant name. Now reload it and make sure that the cliplist name is at the top of the viewfinder. Now whenever you record your clips will get saved into that clip list. Very handy organisation if you are shooting more than one scene on the same disc. By default the RET button is used to add the Shotmark 1 essence mark. Press this just before you stop recording to mark a good clip. You can also go into the menus to set up automatic essence marks so that the camera will automatically mark a clip with peaked audio etc.

Which camera are you using? Like many things in video its a case of 'how long is a piece of string?' There's the usual suspects such as making sure not to blow out highlights, but there are other aspects too. If you are using something like a DVX100 or a higher end camera you will want to be shooting in progressive scan mode. That much is a given. You will also need to adjust the detail circuits in the camera if you are transferring to film. Different DV to film transfer houses have their own preferences as to what camera settings to use. Swiss Effects http://www.swisseffects.ch are very helpful in this regard as are http://www.dvfilm.com

Thats a pretty pedantic argument when a) the main NLE camera rivals are the Panny 800 and the Sony 530. Both SD cameras, and b ) the cost of P2 is such that HD still isn't mainstream viable on those cards. Sony are already working on the HD version of XDCAM. By the time P2 comes down in price by any degree the HD version of XD will be out. Certainly if you were making a HD drama or movie these days you would choose a Varicam or one of the Sony F900 series cameras (or a Viper). Not a P2 camera. So that argument is rather mute.

Hehe, I've seen many put downs of the xdcam system by Panny reps over time. Both companies want to sell their product so its hardly unreasonable that the Sony rep would say that the 530 blows away the SDX just as you defend P2 with all the vigor of a Bengal Tiger. But regardless, the XDCAM cameras and decks hook upto the PC and act like another drive. Just like P2 the MXF files can be dragged straight onto compatible NLEs timelines. You can also edit the proxies and then conform the final output like you mention. The method is up to the user in question. Since I believe P2 also records proxies the workflow can be pretty similar. For example if I am on location for a few days I can edit the proxies on a really low powered and small laptop in the evenings so that when I get back to base I've already got a rough edit of the video done using PDZ1. It'll be the same for P2 when you don't want to be storing the high res video on a field laptop and instead just want to perform a rough edit. Now, as more NLEs get MXF compatibility the PDZ1 software becomes less relevant so that actual, almost finalised edits can be performed on location with a low powered laptop. To me the arguments that go on between P2 and XDCAM are mute and meaningless. One person will prefer one format, and another person will like the other instead. It is all dependant on circumstances and the type of work people do. At the moment the advantages and respective disadvantages of each system balance each other out. For example P2 has the running time and constantly having to upload footage problem, but has faster access times. On the other side of the coin the XDCAM has slower file access when transferring footage., but allows much longer shooting periods before any footage transfer needs to take place. Personally after weighing things up (and the 0% finance on XDCAM helped!) I decided I would much rather have the cheaper discs and longer recording times. A lost or damaged disc is easier to replace than a P2 card (although thats assuming there wasn't any valuable footage on it!) I think the two systems will co-exist with people making similar reasoned descisions based on their working environment/company needs.

That's not really true. The CCD does have a fairly high pixel count, although due to some technical speak that I don't understand, as per all cameras of its type including the SDX900 etc the real pixels used are pure SD resolution. I wish I could remember the reasons and technical speak so perhaps someone else more knowledgable could chime in. It is possible to use an SDI board to take 4:2:2 images straight off the camera head uncompressed on both the 510 and the 530 however, but they will still be SD resolution. One thing of interest is that the Grass Valley Viper Filmstream camera apparently uses the same CCD block as their other cameras including the SD ones (its even mentioned in their Filmstream brochure) But as far as the xdcam goes, no, there's no chance of converting either camera into an HD one. Although Sony are working on a version it will be a few years yet. Enough time for me to save up for a HD lens to take the sting out of buying a camera body.

As anybody here will know the Viper Filmstream camera is a piece of equipment to die for. looking at its specs and it's workflow method got me thinking about the way we obtain images from all cameras. There is even a Reelstream project for the DVX100 to obtain a raw image off the CCDs. Which leads me to camera settings in general if one doesn't have direct access to the raw CCD image if we want to make the most of what we have for post production. Taking the Sony shouldermounts for example, and I would imagine the SDX900, it is possible to install an SDI board and obtain a totally uncompressed 4:2:2 signal. In the field this would of course have to be recorded back to some form of hard drive. But regarding the image adjustment in general, is there anyone here who advocates the turning off of the Matrix and detail functions of their cameras in order to minmise any electronic processing and therefore give an image that can be worked on more in post?